Processes using nucleoside triphosphates with stable aminoxy groups

This is a continuation-in-part of U.S. patent application Ser. No. 16/679,501, filed 11 Nov. 2019, currently pending, which was a continuation in part of U.S. patent application Ser. No. 15/460,475, filed 16 Mar. 2017 now U.S. patent Ser. No. 10/472,383, and U.S. patent application Ser. No. 15/786,086, filed 17 Oct. 2017 now U.S. patent Ser. No. 10/654,841. U.S. patent application Ser. No. 15/786,086 is a continuation in part of U.S. patent application Ser. No. 15/475,694, filed 31 Mar. 2017, and now abandoned. This is also a continuation-in-part of U.S. patent application Ser. No. 16/887,951, filed 29 May 2020, pending. This is also a continuation-in-part of U.S. patent application Ser. No. 16/679,887, filed 29 May 2020, pending.

This invention was made with government support under R41GM119494 awarded by the National Institutes of General Medical Sciences. The government has certain rights in the invention.

Not applicable

None

This invention relates to the field of nucleic acid chemistry, and more specifically to DNA and DNA-like molecules that have a 3′-ONHgroup rather than the 3′-OH group that is found in standard DNA, DNA with non-standard nucleobases, and DNA-like molecules (collectively hereinafter “DNA”). Still more specifically, this invention relates to enzymatic processes that attach, to the end of a standard DNA, RNA, and/or DNA-like or RNA-like molecule, a nucleotide that has its 3′-OH group substantially completely replaced by a 3′-ONHgroup, using a triphosphate analog in an aqueous solution that lacks hydroxylamine. This invention also relates to processes where a DNA polymerase, a reverse transcriptase, a polynucleotide polymerase (e.g. polyA polymerase or polyU polymerase [Church, G. M., Wiegand, D. J., Kohnan, R. E., Kuru, E., Ricttichier, J., Conway, N. (2020) Enzymatic RNA Synthesis, WO/2020/077227 A2][Heinisch, T., Champion, E., Sune, E., Soskine, M. (2021) Template Free Enzymatic synthesis of polynucleotides using Poly(A) and Poly (U) polymerases, WO/2021/018919], or a terminal deoxynucleotide transferase uses a nucleoside triphosphate having a 3′-ONHgroup to add a 3′-aminoxy-2′,3′-dideoxynucleotide to the 3′-end of a DNA or DNA-like molecule.

Well-known in the art are useful processes that require that the enzymatic extension of a DNA, DNA-like, or RNA oligonucleotide (hereinafter a primer) be terminated after introduction of just a single nucleotide at the 3′-end. This extension may be templated, as in the primer-extension processes that are catalyzed by DNA polymerases, RNA polymerases, or reverse transcriptases. Here, successful termination after the addition of just one nucleotide underlies many DNA sequencing architectures, especially those known to use “cyclic reversible termination”. Termination of template-guided extension after the addition of a single nucleotide is frequently achieved by contacting the enzyme to analog of a nucleoside triphosphate where the nucleoside has been altered so as to no longer have a free 3′-hydroxyl group.

Also well known in the art are processes where extension in not templated. Here, a common enzyme to catalyze the process is a terminal deoxynucleotide transferase (TdT). Termination after addition of just a single nucleotide is used in many DNA synthesizing architectures.

Well known among these analogs are triphosphates where the 3′-hydroxyl group is replaced by a hydrogen atom to generate 2′,3′-dideoxynucleoside triphosphates. These are substrates for many polymerases, including many modified polymerases. In forms that carry side chains carrying reporter groups, these have long been used in DNA sequencing processes. Since no convenient method is available to replace the 3′-H by a 3′-OH group on an oligonucleotide, the termination of the oligonucleotide extension process in the presence of a 2′,3′-dideoxynucleoside triphosphate analog is said to be irreversible.

Other nucleoside and oligonucleotide derivatives lacking the standard 3′-OH have functionality that can later be converted to a 3′-OH group under conditions that do not damage oligonucleotides. This allows template-directed primer extension to be terminated “reversibly”.

For example, various patents, including U.S. Pat. Nos. 7,544,794, 8,034,923, and 8,212,020, disclosed that a 3′-O—NHgroup may be used a reversibly terminating moiety. These are referred to as 3′-aminoxy-2′,3′-dideoxynucleosides, -tides, and triphosphates. After a nucleotide having a 3′-O—NHgroup is added to the 3′-end of an oligonucleotide primer, further polymerase-catalyzed extension cannot occur.

This terminating 3′-O—NHgroup may not be removed, allowing its reactivity to be used for a variety of purposes. For example, the 3′-O—NHgroup can react with another molecule that carries an aldehyde or ketone moiety to form useful oximes.

However, if the appropriate reagents are added, the nitrogen-oxygen bond of the 3′-O—NHgroup can be cleaved, thereby converting the 3′-ONHgroup to a 3′-OH group. Once converted, enzymatic extension can proceed. U.S. Pat. Nos. 7,544,792 and 8,212,020 did not provide a practical reagent for cleaving the nitrogen-oxygen bond in the 3′-O—NHunit in an oligonucleotide to generate an extendable 3′-OH group, U.S. Pat. No. 8,034,923 did. U.S. Pat. No. 8,034,923 taught that the nitrogen-oxygen bond in the 3′-O—NHunit could be cleaved by an aqueous solution of sodium nitrite buffered to a pH of near six. The product of that cleavage reaction is a 3′-OH group.

The art is, however, defective when it concerns the 3′-O—NHunit. For the process using this unit to be useful for both DNA sequencing and for enzyme-based DNA synthesis, the triphosphate must be substantially free of contaminant triphosphate having a free 3′-OH group, where “substantially” means in this context less than 0.1%. Preferably, the level of contaminating 3′-OH group is still lower. Otherwise, the enzyme adds a second nucleotide to the DNA molecule that serves as a primer to give an “n+2” product; the n+1 product arises from the desired addition of just one nucleotide. In enzyme synthesis architectures using TdT, this leads to products that are contaminated with analogous products and have a single nucleotide inserted at undesired positions throughout the synthetic product. In enzyme sequencing architectures, n+2 products lead to confusion and sequencing reads.

The significance of this defect should not be overlooked. It is visible in all experiments seen with the triphosphates present in the art prior to the priority date of the instant application. An example is shown in. Here, 0.1% contaminant generates about 3% n+2 product, reflecting the fact that the terminal transferase disproportionately prefers the 3′-OH triphosphate over the more abundant 3′-ONHtriphosphate.

As taught in U.S. patent application Ser. No. 15/460,475 (Benner, S. A. (2017) Nucleoside Triphosphates with Stable Aminoxy Groups. Filed 16 Mar. 2017), for which this is a continuation in part and whose disclosure is incorporated in its entirety herein by reference, the processes disclosed in the prior art for making triphosphates with a 3′-O—NHunit did not provide triphosphates sufficiently free of triphosphates having 3′-OH groups for the processes to have their full utility. Specifically, all the preparations disclosed, intrinsically because of the method of their preparation, contained small but substantial amounts of 3′-nucleoside triphosphates having unblocked 3′-OH groups, here defined as 0.1% or greater. Thus, use of these in template reactions frequently did not lead to termination of all oligonucleotide chains.

As noted in a declaration submitted during the prosecution of patent application Ser. No. 15/460,475, a solid phase process was invented that allowed triphosphate having a 3′-ONHgroup to be synthesized substantially free of triphosphates having a 3′-OH group. This process is based on the unexpected discovery that this outcome required strict exclusion of hydroxylamine from both the manufacturing and use processes of these triphosphates. This process is covered by allowed claims in patent application Ser. No. 15/460,475.

Further, patent application Ser. No. 15/460,475 disclosed compositions of matter that are nucleoside triphosphates of both standard and nonstandard nucleosides having 3′-ONHgroups that are substantially free of triphosphates having a 3′-OH groups. These compositions are also covered by allowed claims in patent application Ser. No. 15/460,475.

The instant invention covers enzymatic processes that use compositions of matter covered by allowed claims in patent application Ser. No. 15/460,475, which are produced by a process covered by allowed claims in patent application Ser. No. 15/460,475. Various of these processes were disclosed in application Ser. No. 15/786,086, whose disclosure is also incorporated herein in its entirety by reference. The instant invention covers enzymatic processes that use compositions of matter covered by allowed claims in patent application Ser. No. 16/887,951, whose disclosure is also incorporated herein in its entirety by reference.

As noted in the description of the prior art, two types of processes benefit from the ability to extend a DNA molecule by just a single nucleotide. In the first type, the extension of the primer is untemplated. In this class of process, the invention disclosed here involves contacting a DNA primer oligonucleotide in an appropriate buffer with an enzyme known as terminal deoxynucleotide transferase (TdT, or simply terminal transferase). Terminal transferase was discovered many years ago in calf thymus as an enzyme that adds nucleoside triphosphates to the 3′-end of an oligonucleotide in an untemplated fashion. Some key references, which are incorporated herein by citation, are:

For standard nucleotides, defined as those that have standard nucleobases such as adenine, guanine, cytosine, and thymine, data disclosed in U.S. Pat. No. 8,034,923 show that terminal transferases accept triphosphates having a 3′-ONHgroup. The presently preferred terminal deoxynucleotidyl transferase (TdT) prefers DNA as an oligonucleotide substrate. Single ribonucleotide addition is seen with the native enzyme a slower rate. The presently preferred TdT is the enzyme that is commercially available, sold by New England Biolabs or Promega, or the analogous enzyme obtained from other mammalian thymus glands. Most preferred is a TdT or one of its variants containing 1-3 amino acid replacements obtained via recombinant DNA technology. However, U.S. Pat. No. 8,034,923 does not constitute anticipatory prior art as it does not disclose triphosphates that are free of hydroxylamine.

The importance of freedom from hydroxylamine is disclosed in Application 154604475. In detail. U.S. Pat. No. 8,034,923 taught that hydroxylamine (HONH) should also be present in compositions containing triphosphates having a 3′-ONHmoiety. U.S. Pat. No. 8,034,923 taught that having hydroxylamine to compositions containing the triphosphates would scavenge adventitious aldehydes, or reversibly cleave the oximes should they be formed.

However, upon storage or even when freshly used, it was discovered that the 3′-ONHunit decomposed to generate a triphosphate carrying a 3′-OH moiety in these prior art preparations. While not wishing to be bound by theory, the conversion of the 3′-nucleoside-O—NHgroup to a 3′-nucleoside-OH appeared to arise from the presence of hydroxylamine, the same hydroxylamine taught in the art (including U.S. Pat. No. 8,034,923) to be used with those triphosphates.

Accordingly, Application 154604475 disclosed compositions of triphosphates that lacked hydroxylamine. These are the compositions used in the herein claimed processes, wherein a key limitation of the claims is the substantial absence of hydroxylamine, defined to be less than one micromolar, more preferably less than one nanomolar. Also claimed in Application 154604475 are inventive processes that deliver such compositions.

Also not disclosed in the prior art was the ability of such compositions to include triphosphates of nucleosides wherein the nucleobases or nonstandard. These are nucleobases other than adenine, guanine, cytosine, and thymine, but rather presented hydrogen bond patterns different from those found in the standard bases. These are shown in, and reduced to practice in the examples.

With respect to untemplated primer extension by just one nucleotide, the invention here comprises contacting an oligonucleotide, preferably an oligo-2′-deoxyribonucleotide, with a triphosphate, as disclosed in U.S. Pat. No. 8,034,923, carrying the 3′-ONHmoiety, in aqueous buffers where TdT operates. These buffers are well known in the art, and are provided in the examples below. The buffer may optionally contain divalent cobalt cation (Co), which may improve the ability of the terminal transferase to accept standard pyrimidine nucleoside triphosphates. However, with the aminoxy analogs, we have discovered that Codoes not improve the performance of TdT, at least in the buffers examined. Those buffers have preferable pH ranges from 7 to 8, but not outside pH 6 to 9. The preferable contact temperature is preferably between 25° C. and 40° C.

The utility of the instant invention arises from its ability to at a single nucleotide at a time.

This can be, for example, envisioned as a synthesis procedure, where an oligonucleotide having a defined, preselected, sequence is synthesized by contacting an immobilized primer with:

For this and other applications where addition of just one nucleotide is desired, is preferred that the triphosphate having a 3′-ONHmoiety not be contaminated with triphosphates that have a standard, and extendable, 3′-OH group. As shown in the examples, TdT has a preference for the natural triphosphate having an extendable 3′-OH group. Indeed, TdT can be used to clean up preparations of triphosphates having a 3′-ONHmoiety by removing natural triphosphates having an extendable 3′-OH group. For other applications, this contamination is tolerable.

The most presently preferred triphosphates having a 3′-ONHmoiety are those prepared by the procedure disclosed in U.S. patent application Ser. No. 15/460,475, which is incorporated herein in its entirety by reference. These nucleoside triphosphates are substantially free of contaminating standard triphosphate, where “substantially” means that the preparation triphosphate carrying a 3′-O—NHmoiety contains less than 0.5 mole percent of the analogous triphosphate with a free 3′-OH group, more preferably less than 0.05 mole percent, and most preferably less than 0.005 mole percent, calculated relative to the triphosphate carrying a 3′-O—NHmoiety. Further, as taught in U.S. patent application Ser. No. 15/460,475, in addition to the standard nucleoside triphosphates shown in, the nucleoside might carry an unnatural nucleobase, selected from the structures shown in.

The art (for example Hutter et al. 2010) shows that aminoxy derivatives can be used in templated-directed oligonucleotide primer extension using DNA polymerases. Here, the oligonucleotide primer is presented with a template that contains a segment that is substantially complementary to the primer, where the primer is hybridized to the template, where the template has a single stranded segment 5′-distal to the segment where the primer is hybridized. The first nucleotide in this single stranded segment directs the polymerase to incorporate the 3′-aminoxy triphosphate that is singly added by the polymerase. This primer-template can be a hairpin.

The presently preferred polymerase is that from, or a mutant thereof, as disclosed in Hutter et al. [2010]. However, neither Hutter et al. nor any other publication prior to the instant priority constitutes anticipatory prior art, as it does not disclose preparations of triphosphates that are free of hydroxylamine.

Optionally, the solution may also contain an alkoxylamine in addition to the alkoxylamine that is the triphosphate itself. Preferably, the additional dissolved alkoxylamine is a lower alkoxylamine such as methoxylamine. Preferably, the concentration of the alkoxylamine is between 10 and 50 micromolar, and is delivered in a sequence where the enzyme is first contacted with the primer, then contacted with the additional alkoxylamine, and then contacted with the triphosphates. This was discovered to prevent reaction of alkoxylamine with cytosine.

The solution to the stability problem opened up a wide range of hitherto unanticipated applications. For example, Karalkar and Benner solved the difficult problem of synthesizing the analogous ribonucleoside derivatives, where Z is OH or P-Me in the formula below (see U.S. patent application Ser. No. 16/887,951, filed 29 May 2020, incorporated in detail, including figures, legends, specification, abstract, examples, and claims].

This extends the range of useful enzymes to DNA polymerases, reverse transcriptases, polynucleotide polymerases (e.g. polyA polymerase, polyU polymerase or their variants [Church, G. M., Wiegand, D. J., Kohnan, R. E., Kuru, E., Ricttichier, J., Conway, N. (2020) Enzymatic RNA Synthesis, WO/2020/077227 A2][Heinisch, T., Champion, E., Sune, E., Soskine, M. (2021) Template Free Enzymatic synthesis of polynucleotides using Poly(A) and Poly (U) polymerases, WO/2021/018919], and terminal deoxynucleotide transferases, and includes both templated and untemplated syntheses. This includes variants of these generated by site directed mutagenesis, directed evolution, screening, or other processes known to the artisans of diversity science. Also applicable are variants of polymerases (e.g., polymerase theta) [Pomerantz, R. T. (2016) Modification of 3′ terminal ends of nucleic acids by DNA polymerase theta. WO2017075421A1], which provided processes for modifying the 3′-terminal ends of nucleic acids using DNA polymerase 0 terminal transferase activity.

The ability of terminal transferase to add a 3′-aminoxy terminating triphosphate to oligonucleotides was discovered by a series of experiments. In these experiments, this oligonucleotide substrate was used.

This oligonucleotide was 5′-labeled to give

which is SEQ ID NO 1) using OptiKinase and gamma-labeled radioactive ATP.

Two different buffers were used for the experiments that discovered the ability of terminal transferase to add a 3′-aminoxy terminating triphosphate to oligonucleotides. The “purine tailing buffer” contained 100 mM cacodylate buffer (pH 7.1), 2 mM MnCl, 0.1 mM DTT, 10 pmol of radiolabeled template (0.5 μM), 10 units of Terminal Transferase, and varying amounts of reversible terminating triphosphates ranging from 5 μM to 250 μM (as indicated on gel). The total volume was 20 μL.

The “pyrimidine tailing buffer” contained 100 mM cacodylate buffer (pH 7.1), 2 mM CoCl, 0.1 mM DTT, 10 pmol of radiolabeled template (0.5 μM), 10 units of Terminal Transferase, and varying amounts of reversible terminator triphosphates from 5 μM to 250 μM. Again, the total volume was 20 μL.

Samples were incubated at 37° C. for 1 hour. Then, the transferase reaction was terminated by heating at 70° C. for 10 min. Loading buffer 10 μL (98% formamide, 10 mM EDTA 1 mg/mL xylene cyanol and 1 mg/mL bromophenol blue) was added to each reaction mixture, and an aliquot containing 2 pmoles of products (4 μL) was resolved on 8% PAGE. An additional study was done testing just G-ONHat 250 μM at 2, 5, 15, 30 and 60 min incubation at 37° C.

Data are shown in. Terminal transferase was able to incorporate all four nucleoside triphosphates containing the 3-ONHblock. In the purine buffer, dGTP-ONHappears to be incorporated more easily than dATP-ONH. However even though this buffer is sold commercially to prefer standard purine nucleosides over pyrimidine nucleosides, dTTP-ONHis still incorporated. However, in purine buffer, dCTP-ONHis only slightly incorporated, and appears to inhibit the reaction at higher concentrations.

In the pyrimidine buffer, dGTP-ONHand dATP-ONHboth are successfully incorporated, as is dTTP-ONH. dCTP-ONHis perhaps incorporated better, but again appears to have inhibitory activity.

The results inwere obtained with triphosphates containing a small amount of 3-OH unblocked nucleoside triphosphates. The bands indicating incorporation of more than one nucleotide indicate this.

A time course for the incorporation of GTP-ONHusing TdT was then determined (). Nearly all of the primer has acquired an additional oligonucleotide) is seen. Thus, contacting oligonucleotide with the triphosphates is shown to be a way to incorporate the reversible terminator without the need of a template.

The oligonucleotide primer used here is the same as in Example 1. The aminoxytriphosphates were prepared on a resin, as taught in U.S. patent application Ser. No. 15/460,475, but under a release procedure using HONHelution, which did not minimize the presence of contaminating triphosphates with a free 3′-OH group. This shows the results of this procedure. Data are shown in. The lanes are as follows:

The TdT reaction was carried out as before withP-labeled primer (0.5 μM) in 1× in terminal transferase buffer (20 mM Tris-acetate, pH 7.9, 50 mM potassium acetate, 10 mM magnesium acetate, 1 mM dithiotheritol), 0.25 mM CoCl, 200 μM reversible terminator samples (a-f) and 10 Units of terminal deoxynucleotidyl transferase (TdT). Reactions were incubated at 37° C. for five and 50 min. Reactions were quenched by the addition of formamide quench buffer and were resolved on a 20% PAGE ().

The TdT reaction was executed with dTTP-ONHprepared as taught in U.S. patent application Ser. No. 15/460,475 using the same primer as above. Here, the aminoxytriphosphates were prepared by releasing TTP-ONHfrom the resin using MeONH. Data are shown in.

As before,P labeled primer (0.5 μM) in 1× in terminal transferase buffer (20 mM Tris-acetate, pH 7.9, 50 mM potassium acetate, 10 mM magnesium acetate, 1 mM dithiotheritol), with and without 0.25 mM CoCl, 200 μM reversible terminator samples (a-d) and 10 Units of terminal deoxynucleotidyl transferase (TdT). Control samples containing TTP and ddTTP were also tested. Reactions were incubated at 37° C. 5 and 60 min. Reactions were quenched by the addition of formamide quench buffer and were resolved on a 20% PAGE ().

1) Terminal Transferase 3′-Tailing Studies